From f4e79b105d5162b6d28f302069f8305dfb17c063 Mon Sep 17 00:00:00 2001
From: david dali susanibar arce <davi.sarces@gmail.com>
Date: Tue, 13 Sep 2022 19:48:38 -0500
Subject: [PATCH] ARROW-17418: [Doc][Java] Dataset library able to compile with
 mvn command (#13889)

Dataset/ORC/Gandiva library (.so / .dylib) able to compile with mvn command

$ mvn clean generate-resources -Pgenerate-jnicpp-dylib_so -DARROW_DATASET=ON -DARROW_ORC=ON -DARROW_GANDIVA=ON -N

Authored-by: david dali susanibar arce <davi.sarces@gmail.com>
Signed-off-by: Sutou Kouhei <kou@clear-code.com>
---
 docs/source/developers/java/building.rst |  27 +++-
 java/pom.xml                             | 191 +++++++++++++++++++++++
 2 files changed, 214 insertions(+), 4 deletions(-)

diff --git a/docs/source/developers/java/building.rst b/docs/source/developers/java/building.rst
index fd125e178db53..42faadb8d219d 100644
--- a/docs/source/developers/java/building.rst
+++ b/docs/source/developers/java/building.rst
@@ -100,7 +100,7 @@ for that architecture. See, for example, the `Azul JDK <https://www.azul.com/dow
 Maven
 ~~~~~
 
-- To build only the C Data Interface library:
+- To build only the JNI C Data Interface library:
 
     .. code-block::
 
@@ -111,9 +111,28 @@ Maven
         $ ls -latr ../java-dist/lib
         |__ libarrow_cdata_jni.dylib
 
-- To build other JNI libraries:
+- To build JNI ORC & JNI Gandiva libraries:
+
+    .. code-block::
+
+        $ cd arrow/java
+        $ export JAVA_HOME=<absolute path to your java home>
+        $ java --version
+        $ mvn clean generate-resources -Pgenerate-jnicpp-dylib_so -N
+        $ ls -latr  java-dist/lib
+        |__ libarrow_orc_jni.dylib
+        |__ libgandiva_jni.dylib
+
+- To build only the JNI Dataset library:
 
-    Not options available on Maven.
+    .. code-block::
+
+        $ cd arrow/java
+        $ export JAVA_HOME=<absolute path to your java home>
+        $ java --version
+        $ mvn clean generate-resources -Pgenerate-dataset-dylib_so -N
+        $ ls -latr java-dist/lib
+        |__ libarrow_dataset_jni.dylib
 
 CMake
 ~~~~~
@@ -175,7 +194,7 @@ CMake
         |__ libarrow_orc_jni.dylib
         |__ libgandiva_jni.dylib
 
-- To build only the Dataset library:
+- To build only the JNI Dataset library:
 
     .. code-block::
 
diff --git a/java/pom.xml b/java/pom.xml
index 7bdfa7c656341..8abe7ad9dcaa9 100644
--- a/java/pom.xml
+++ b/java/pom.xml
@@ -969,6 +969,197 @@
       </build>
     </profile>
 
+    <profile>
+      <id>generate-dataset-dylib_so</id>
+      <properties>
+        <arrow.dataset.jni.dist.dir>java-dist/lib</arrow.dataset.jni.dist.dir>
+        <cpp.dependencies.builded>false</cpp.dependencies.builded>
+      </properties>
+      <build>
+        <plugins>
+          <plugin>
+            <groupId>org.codehaus.mojo</groupId>
+            <artifactId>exec-maven-plugin</artifactId>
+            <version>3.1.0</version>
+            <executions>
+              <execution>
+                <id>datasetdir</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>mkdir</executable>
+                  <commandlineArgs>-p cpp-jni java-dist java-jni</commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+              <execution>
+                <id>datasetarrowdependency</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    -S cpp
+                    -B cpp-jni
+                    -DARROW_CSV=ON
+                    -DARROW_DATASET=ON
+                    -DARROW_DEPENDENCY_SOURCE=BUNDLED
+                    -DARROW_DEPENDENCY_USE_SHARED=OFF
+                    -DARROW_FILESYSTEM=ON
+                    -DARROW_GANDIVA=ON
+                    -DARROW_GANDIVA_JAVA=ON
+                    -DARROW_GANDIVA_STATIC_LIBSTDCPP=ON
+                    -DARROW_JNI=ON
+                    -DARROW_ORC=ON
+                    -DARROW_PARQUET=ON
+                    -DARROW_PLASMA=ON
+                    -DARROW_PLASMA_JAVA_CLIENT=ON
+                    -DARROW_S3=ON
+                    -DARROW_USE_CCACHE=ON
+                    -DCMAKE_BUILD_TYPE=Release
+                    -DCMAKE_INSTALL_LIBDIR=lib
+                    -DCMAKE_INSTALL_PREFIX=java-dist
+                    -DCMAKE_UNITY_BUILD=ON
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                  <skip>${cpp.dependencies.builded}</skip>
+                </configuration>
+              </execution>
+              <execution>
+                <id>datasetarrowdependencybuild</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    --build cpp-jni --target install --config Release
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                  <skip>${cpp.dependencies.builded}</skip>
+                </configuration>
+              </execution>
+              <execution>
+                <id>datasetdefine</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    -S java
+                    -B java-jni
+                    -DARROW_JAVA_JNI_ENABLE_DATASET=ON
+                    -DARROW_JAVA_JNI_ENABLE_DEFAULT=OFF
+                    -DBUILD_TESTING=OFF
+                    -DCMAKE_BUILD_TYPE=Release
+                    -DCMAKE_INSTALL_PREFIX=${arrow.c.jni.dist.dir}
+                    -DCMAKE_PREFIX_PATH=${project.basedir}/../java-dist
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+              <execution>
+                <id>datasetbuild</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    --build java-jni --target install --config Release
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+            </executions>
+          </plugin>
+        </plugins>
+      </build>
+    </profile>
+
+    <profile>
+      <id>generate-jnicpp-dylib_so</id>
+      <build>
+        <plugins>
+          <plugin>
+            <groupId>org.codehaus.mojo</groupId>
+            <artifactId>exec-maven-plugin</artifactId>
+            <version>3.1.0</version>
+            <executions>
+              <execution>
+                <id>jnidir</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>mkdir</executable>
+                  <commandlineArgs>-p java-dist cpp-jni</commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+              <execution>
+                <id>jnidefine</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    -S cpp
+                    -B cpp-jni
+                    -DARROW_CSV=ON
+                    -DARROW_DATASET=ON
+                    -DARROW_DEPENDENCY_SOURCE=BUNDLED
+                    -DARROW_DEPENDENCY_USE_SHARED=OFF
+                    -DARROW_FILESYSTEM=ON
+                    -DARROW_GANDIVA=ON
+                    -DARROW_GANDIVA_JAVA=ON
+                    -DARROW_GANDIVA_STATIC_LIBSTDCPP=ON
+                    -DARROW_JNI=ON
+                    -DARROW_ORC=ON
+                    -DARROW_PARQUET=ON
+                    -DARROW_PLASMA=ON
+                    -DARROW_PLASMA_JAVA_CLIENT=ON
+                    -DARROW_S3=ON
+                    -DARROW_USE_CCACHE=ON
+                    -DCMAKE_BUILD_TYPE=Release
+                    -DCMAKE_INSTALL_LIBDIR=lib
+                    -DCMAKE_INSTALL_PREFIX=java-dist
+                    -DCMAKE_UNITY_BUILD=ON
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+              <execution>
+                <id>jnidefinebuild</id>
+                <phase>generate-resources</phase>
+                <goals>
+                  <goal>exec</goal>
+                </goals>
+                <configuration>
+                  <executable>cmake</executable>
+                  <commandlineArgs>
+                    --build cpp-jni --target install --config Release
+                  </commandlineArgs>
+                  <workingDirectory>../</workingDirectory>
+                </configuration>
+              </execution>
+            </executions>
+          </plugin>
+        </plugins>
+      </build>
+    </profile>
+
   </profiles>
 
 </project>